mlsanityData to 14 Oct 2026

A step towards understanding why classification helps regression

arXiv 2308.10603 ↗First posted 21 Aug 2023cs.CVcited by 15

Not on the map. Why?

Abstract (arXiv)

A number of computer vision deep regression approaches report improved results when adding a classification loss to the regression loss. Here, we explore why this is useful in practice and when it is beneficial. To do so, we start from precisely controlled dataset variations and data samplings and find that the effect of adding a classification loss is the most pronounced for regression with imbalanced data. We explain these empirical findings by formalizing the relation between the balanced and imbalanced regression losses. Finally, we show that our findings hold on two real imbalanced image datasets for depth estimation (NYUD2-DIR), and age estimation (IMDB-WIKI-DIR), and on the problem of imbalanced video progress prediction (Breakfast). Our main takeaway is: for a regression task, if the data sampling is imbalanced, then add a classification loss.

Key facts. Cited by 15 papers in the map (data through 14 Oct 2026). Builds on 34 of its 52 references that are in the map. First posted on arXiv 21 Aug 2023.

Builds on

34 of 52 references.

References · 34

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2015Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networkscited by 9,639
  3. 2017Mask R-CNNcited by 7,277
  4. 2015SSD: Single Shot MultiBox Detectorcited by 3,130
  5. 2015Fast R-CNNcited by 2,933
  6. 2019Objects as Pointscited by 1,175
  7. 2017Focal Loss for Dense Object Detectioncited by 752
  8. 2018Deep Ordinal Regression Network for Monocular Depth Estimationcited by 723
  9. 2021Deep Long-Tailed Learning: A Surveycited by 417
  10. 2017End-to-End Learning of Geometry and Context for Deep Stereo Regressioncited by 360
  11. 20163D Bounding Box Estimation Using Deep Learning and Geometrycited by 278
  12. 2016Unsupervised Feature Extraction by Time-Contrastive Learning and Nonlinear ICAcited by 273
22 more
  1. 2017A General and Adaptive Robust Loss Functioncited by 236
  2. 2021Delving into Deep Imbalanced Regressioncited by 158
  3. 2020Overcoming Classifier Imbalance for Long-tail Object Detection with Balanced Group Softmaxcited by 152
  4. 2018Revisiting Single Image Depth Estimation: Toward Higher Resolution Maps with Accurate Object Boundariescited by 145
  5. 2020Self-supervised Video Representation Learning by Pace Predictioncited by 139
  6. 2021TCLR: Temporal Contrastive Learning for Video Representationcited by 136
  7. 2015Render for CNN: Viewpoint Estimation in Images Using CNNs Trained with Rendered 3D Model Viewscited by 94
  8. 2022Balanced MSE for Imbalanced Visual Regressioncited by 89
  9. 2019Hyperspherical Prototype Networkscited by 86
  10. 2019Unsupervised learning of action classes with continuous temporal embeddingcited by 76
  11. 2014Designing Deep Networks for Surface Normal Estimationcited by 59
  12. 2014Viewpoints and Keypointscited by 59
  13. 2022RankSim: Ranking Similarity Regularization for Deep Imbalanced Regressioncited by 52
  14. 2018A Comprehensive Analysis of Deep Regressioncited by 48
  15. 2019Pose from Shape: Deep Pose Estimation for Arbitrary 3D Objectscited by 39
  16. 2015Dense Optical Flow Prediction from a Static Imagecited by 36
  17. 2016Horizon Lines in the Wildcited by 30
  18. 2015Robust Optimization for Deep Regressioncited by 22
  19. 2016Subcategory-aware Convolutional Neural Networks for Object Proposals and Detectioncited by 17
  20. 2017Incremental Boosting Convolutional Neural Network for Facial Action Unit Recognitioncited by 16
  21. 2019Monocular Pedestrian Orientation Estimation Based on Deep 2D-3D Feedforwardcited by 3
  22. 2020Classification Beats Regression: Counting of Cells from Greyscale Microscopic Images based on Annotation-free Training Samplescited by 3

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 6; 2025: 4; 2026: 5.

15 papers.

Sort:

By area:

  1. 2024Stop Regressing: Training Value Functions via Classification for Scalable Deep RLcited by 144
  2. 2024Hierarchical Classification Auxiliary Network for Time Series Forecastingcited by 26
  3. 2024FoldToken: Learning Protein Language via Vector Quantization and Beyondcited by 23
  4. 2024An Efficient Model-Agnostic Approach for Uncertainty Estimation in Data-Restricted Pedometric Applicationscited by 3
  5. 2025Learning Relative Gene Expression Trends from Pathology Images in Spatial Transcriptomicscited by 3
  6. 2024V-Hands: Touchscreen-based Hand Tracking for Remote Whiteboard Interactioncited by 2
  7. 2025Improving Deep Regression with Tightnesscited by 2
  8. 2024Structural Entropy Guided Probabilistic Codingcited by 1
  9. 2025Balanced Sharpness-Aware Minimization for Imbalanced Regressioncited by 1
  10. 2025CardiacCLIP: Video-based CLIP Adaptation for LVEF Prediction in a Few-shot Mannercited by 0
  11. 2026A Latent Risk-Aware Machine Learning Approach for Predicting Operational Success in Clinical Trials based on TrialsBankcited by 0
  12. 2026Rare-Aware Autoencoding: Reconstructing Spatially Imbalanced Datacited by 0
3 more
  1. 2026How Should Transformers Encode Numeric Values in Electronic Health Records?cited by 0
  2. 2026Error Aware Distribution Prediction for Lightweight Implicit Neural Representationscited by 0
  3. 2026PhyRestore: Physics-Structured Latent-Factor Restorationcited by 0